公开(公告)号：EP3049962A1

公开(公告)日：2016-08-03

申请号：EP13894295.8

申请日：2013-09-27

申请人： Intel Corporation

发明人： HUANG, Yicong ,  CHOW, Kingsum

IPC分类号： G06F17/30

CPC分类号： G06F17/30306 ,  G06F7/02 ,  G06F11/3037 ,  G06F11/3433 ,  G06F17/3007 ,  G06F17/30132 ,  G06F17/30221 ,  G06F17/3053 ,  G06F17/30598 ,  G06F17/30864 ,  G06F17/30943

摘要： A mechanism is described for facilitating dynamic data management for computing devices according to one embodiment. A method of embodiments, as described herein, includes tracking one or more factors relating to a plurality of data sets, evaluating the plurality of data sets based on the one or more factors. The evaluating may include speculating at least one of relevancy and accessibility of each of the plurality of data sets. The method may further include generating data scores, the data scores being associated with the plurality of data sets based on the evaluation of the plurality of data sets, performing a first comparison of the data scores of the plurality of data sets with a criteria score, and classifying each data set based on the first comparison. The classifying may include setting caching order for each data set of the plurality of data sets.

摘要翻译： 一种机制描述了一种用于促进动态数据管理用于计算设备的gemäß一个实施例。 实施例的方法中，如所描述的，包括跟踪与数据集的多个一种或多种因素的影响，评估基于所述一个或多个因素的数据集的多元性。 该评估可以包括推测每个数据集的所述多个的相关性和可访问性中的至少一个。 该方法可以进一步包括生成数据的分数，所述分数数据被基于数据集的所述多个评估数据集的所述多个相关联，在执行数据集的所述多个与标准评分的数据分数的第一比较， 和分类基于所述第一比较每个数据集。 所述分类可以包括设置为每个数据集的数据集的所述多个高速缓存顺序。